ACIT4630 Advanced Machine Learning and Deep Learning Course description

This course provides broad introduction to machine learning which includes supervised, unsupervised, reinforcement learning and deep learning, that can be used in

different application domains. The course will focus on case studies and examples, depending on student interests. Students will learn from studying, presenting and discussing on relevant research articles and expose themselves in research by doing a research project.

Bachelor level knowledge of the following topics is helpful for understanding some of the concepts in this course:

• linear algebra
• vector calculus
• basic statistics and probability.

Some experience with programming, especially with Python, will be beneficial for the project.

No formal requirements over and above the admission requirements.

On successful completion of the course, students should have the following learning outcomes defined in terms of knowledge, skills and general competence.

Knowledge

The student:

• is knowledgeable about supervised, unsupervised, reinforcement learning
• has a good understanding of the principles of state-of-the-art deep neural networks such as CNN, RNN, GAN, RL.
• has a good understanding of both theoretical and practical know how required to use machine and deep learning methods effectively

Skills

The student:

• develop practical skills necessary to build machine learning and deep learning neural networks
• is able to analyze machine learning methods in regard to their performance and effectiveness
• is able to use existing deep learning networks, improve and/or customize them to apply to new problems

General competence

The student:

• has both theoretical and practical understanding of machine and deep learning methods
• can discuss relevance, strength and limitations of machine learning and deep learning in solving real world problems
• is able to work on relevant research projects

This course covers the fundamental principles of machine learning and deep learning methods and best practices in solving problems effectively. Most of the problems are related and applicable in many areas such as computer vision, surveillance, assistive technology, medical imaging etc. Therefore, the course intends to provide case studies and examples of ML and DL in solving various problems. Students can explore tremendous potential of modern AI, ML and DL methods and techniques in solving problems in different application domains.

The course consists of lectures by teaching staff and other guests (PhD students, internal staff, invited external guests), a series of seminars, and a project work. Students will actively participate in the seminars by presenting papers, listening and discussing on other presentations. The aim of the course is to provide a research-based education in the field. Students will do a individual research project with the aim of cultivating them towards good future researchers.

Practical training

None.

The following required coursework must be approved before the student can take the exam:

• Minimum 80% attendance
• 2 individual oral presentations (one on a given topic, one on topic of own choice), and participate as a prepared opponent/discussant for the assigned other presentations.

The assessment will be based on:

• One individual project work implementation and report (about 25 to 35 pages)
• Individual oral exam (about 30 minutes).

Each of them carries about 50% weightage in the final grade. The oral examination cannot be appealed.

Both exams must be passed in order to pass the course.

None.

For the final assessment a grading scale from A to E is used, where A denotes the highest and E the lowest pass grade, and F denotes a fail.

Two sensors, the first sensor from the teaching staff and the second sensor can be either internal or external.